Reply to Comments by Drs Glue and Menkes
The comments and questions raised by Drs Glue and Menkes are conceptually and clinically critical.1 The nub of Drs Glue and Menkes' remarks, which we agree with, is that it remains to be determined, with minimal equivocation, whether the antisuicide effects reported in association with ketamine treatment are a direct affect or an indirect effect (ie, epiphenomenon) coincident with symptom mitigation of depression. Conceptually, depressive symptoms may be disaggregated across multiple dimensions subserved by overlapping yet discrete pathogenetic substrates. A working assumption in depression has been that suicidal ideation, nonlethal self-harm, and lethal self-harm are inextricably intertwined with negative cognitive-emotional states. Empirical evidence, however, indicates that ideation and self-harm gestures are dissociable from negative cognitive-emotional states.
Our interpretation of the articles published regarding ketamine's effect on depressive symptoms is that there is reason to believe that the antisuicide effect may be independent of improvements in depressive symptoms (ie, pseudospecificity vs specificity). In our article, we purposefully emphasize the uncertainty of this observation, which is in large part due to the insufficient methodology; extant studies demonstrating antisuicide effects would require a sufficient and adequate statistical approach (eg, path analysis) and/or an analysis confined to a population without mental illness (eg, major depressive disorder).2 To our knowledge, such a study has not yet been conducted. Our review of the literature indicated that ketamine affects cognitive-emotional processing, which may result in an antisuicide effect not correlated with symptom reduction as measured by conventional depressive symptom metrics.
The plausibility that ketamine could have antisuicide effects independent of its effects on depressive symptoms is speculative but could relate to procognitive properties. We proffered the notion that, via salutary effects on molecular and cellular substrates and neural circuits subserving cognitive function (eg, mTORC1), ketamine may reduce impulsivity and augment executive functions.2 The foregoing is, however, a testable hypothesis (ie, that ketamine reduces suicidality independent of affective symptom mitigation via procognitive effects and awaits empirical confirmation. The plausibility that a procognitive agent may reduce impulsive self-harm is provided by published articles regarding psychostimulants, wherein there has been reduction in motor vehicle accidents and trauma-related admissions in the general population.3,4 Moreover, lithium and clozapine, 2 other agents with antisuicide properties, are also known to exert anti-impulsivity effects, which may be in part due to their beneficial effects on cognitive architectural substrates.5
Interventional and observational studies with ketamine should provide opportunity to further evaluate potential direct antisuicide effects. Sufficient sampling would allow for the appropriate statistical approach to parcellate direct effects. We agree with the notion that studying ketamine across disparate populations would be of clinical and conceptual interest (eg, borderline personality disorder). In the interim, we agree that the antisuicide effects ascribed to ketamine in populations with major depressive disorder and imminent suicide risk cannot be taken as prima facie evidence of direct effect until the appropriate approach has been taken. We would also appeal to colleagues to additionally evaluate cognitive functions in individuals receiving ketamine treatment, which may inform possible mechanisms mediating symptom relief with this agent, as well as possible antisuicide effects.